Summary
Keywords
References
Article Info
Related Articles
Summary
Functional cure of hepatitis B is defined as sustained undetectable circulating HBsAg and HBV DNA after a finite course of treatment. Barriers to HBV cure include the reservoirs for HBV replication and antigen production (covalently closed circular DNA [cccDNA] and integrated HBV DNA), the high viral burden (HBV DNA and HBsAg) and the impaired host innate and adaptive immune responses against HBV. Current HBV therapeutics, 1 year of pegylated-interferon-α (PEG-IFNα) and long-term nucleos(t)ide analogues (NUCs), rarely achieve HBV cure. Stopping NUC therapy may lead to functional cure in some Caucasian patients but rarely in Asian patients. Switching from a NUC to IFN after HBV DNA suppression increases the chance of HBsAg clearance mainly in those with low HBsAg levels. Novel antiviral strategies that inhibit viral entry, translation and secretion of HBsAg, modulate capsid assembly, or target cccDNA transcription/degradation have shown promise in clinical trials. Novel immunomodulatory approaches including checkpoint inhibitors, metabolic modulation of T cells, therapeutic vaccines, adoptive transfer of genetically engineered T cells, and stimulation of innate and B-cell immune responses are being explored. These novel approaches may be further combined with NUCs or PEG-IFNα in personalised strategies, according to virologic and disease characteristics, to maximise the chance of HBV cure. The development of curative HBV therapies should be coupled with the development of standardised and validated virologic and immunologic assays to confirm target engagement and to assess response. In addition to efficacy, curative therapies must be safe and affordable to meet the goal of global elimination of hepatitis B.
Keywords
Background
Despite the availability of safe and effective vaccines for forty years, chronic HBV infection remains a serious threat to global public health, affecting approximately 257 million people worldwide. In June 2016, the World Health Organization published the first global health sector strategy on viral hepatitis, with the goal of reducing the incidence of viral hepatitis by 90% and its associated mortality by 65%, by 2030. In addition to ensuring universal vaccination of all newborns, and improving diagnosis and linkage to care, much effort has been devoted to developing a cure for chronic hepatitis B.
Definitions and clinical implication of functional cure of HBV
Functional cure of HBV
Sterilising HBV cure with elimination of both covalently closed circular DNA (cccDNA) and integrated HBV DNA is unlikely to be feasible in the foreseeable future. The consensus of experts is to aim for functional HBV cure, defined as the sustained loss of detectable HBsAg and HBV DNA in serum, after a finite course of treatment. [4]
HBsAg seroclearance is an indication of marked suppression of HBV replication and cccDNA transcription but HBV is still present in the liver as transcriptionally inactive cccDNA or integrated HBV DNA, and reactivation of HBV replication can occur upon immune suppression. Nonetheless, HBsAg seroclearance is associated with an additional clinical benefit – a further reduction in risk of hepatocellular carcinoma – beyond that of HBV DNA suppression without HBsAg seroclearance. It also allows for the discontinuation of nucleos(t)ide analogues (NUCs) with very low likelihood of seroreversion and viral relapse. From the patients’ perspective, HBsAg seroclearance is a major milestone, removing the stigma that can limit social life and occupational opportunities.
Partial cure of HBV
Partial HBV cure, defined as detectable HBsAg but persistently low or undetectable HBV DNA in serum after completing a course of antiviral therapy, has been proposed as an intermediate step towards HBV cure. Patients who achieve partial HBV cure have normal liver enzymes and inactive liver disease, and favourable clinical outcomes compared to their viraemic counterparts, but outcomes are inferior compared to those in patients who achieve functional cure. This inactive state represents disease remission rather than cure because reactivation can occur either spontaneously or following immunosuppression.
Barriers to HBV cure
The major barriers to HBV cure include the reservoirs for HBV replication and antigen production (cccDNA and integrated HBV DNA), the high viral burden (HBV DNA and HBsAg) and the impaired host innate and adaptive immune responses against HBV. Episomal cccDNA is a key intermediate in the HBV life cycle, serving as a transcriptional template for all HBV RNAs responsible for DNA replication and antigen production. This minichromosome is located within the nucleus of infected hepatocytes, protected from host immune responses and not targeted by NUCs. The transcriptional activity of cccDNA is responsible for ongoing antigen production during NUC therapy and for virologic relapse following NUC withdrawal. cccDNA persists following HBsAg clearance and is responsible for HBV reactivation in patients with resolved HBV infection following rituximab treatment or following transplantation of livers from anti-HBc-positive donors into HBV-naïve recipients.[9], [10]
Any effective strategy for functional cure will require therapies that can directly target cccDNA, through silencing of transcriptional activity, mutagenesis, or degradation. Gene editing with CRISPR Cas-9 could directly degrade cccDNA (by targeting cccDNA genes) or indirectly silence cccDNA transcriptional activity (by targeting the cccDNA-associated histones). Gene editing studies in preclinical models have demonstrated significant reductions in cccDNA activity. However, cure of chronic HBV infection would require 100% editing efficiency in infected hepatocytes, which is not yet achievable with current liver-targeting delivery systems. Earlier studies suggested that the half-life of cccDNA is long, years or decades, and elimination of cccDNA relies on turnover of infected hepatocytes. However, a recent study showed that the half-life of cccDNA may be shorter, weeks or months, suggesting that elimination of cccDNA may be feasible. The shorter estimated half-life of cccDNA in this study is in part related to the realisation that NUCs do not suppress HBV DNA replication completely. Additional studies are needed to confirm this new finding.
Another reservoir for HBV antigen production is integrated HBV DNA. Integration occurs early in the natural history of chronic HBV Infection and increases with duration of infection, although there is recent evidence that this trend may be partly reversed by long-term viral suppression.During the early phase of HBV infection (HBeAg positive), >95% of HBsAg is derived from cccDNA transcripts, whilst in the late phase of infection (HBeAg negative), >50% of HBsAg is derived from integrated HBV DNA. Thus, any HBV cure will need to target both cccDNA and integrated HBV DNA.[18]
Another major barrier to HBV functional cure is the presence of a dysfunctional innate and adaptive immune response to the virus. The high antigen burden, particularly HBsAg, in chronic HBV infection creates a tolerogenic intrahepatic and extrahepatic environment. HBV produces a huge excess of spherical and filamentous subviral particles (SVPs), which contain HBsAg but lack a nucleocapsid and are therefore non-infectious. The high concentration of circulating HBsAg is thought to promote HBV persistence via several mechanisms: acting as a decoy for neutralising hepatitis B surface antibody; downregulating HBsAg- and HBeAg-specific T-cell immunity and suppressing innate immunity through dendritic cell and natural killer cell dysfunction. [21], [22]
Inhibition of HBV DNA replication and HBsAg production may suffice to restore immune responses to HBV in some patients but additional immunomodulatory therapies may be needed for others.
Approaches to achieve functional cure of HBV
Current treatment approaches (Table 1)
Currently available treatment for chronic HBV infection includes 1 year of pegylated-interferon α (PEG-IFNα) and long-term NUCs. Both approaches result in very low rates of HBsAg seroclearance. A 1-year course of PEG-IFNα results in an overall HBsAg seroclearance rate of 2-3% at the end of treatment, increasing to 3-8% after 3 years of post-treatment follow-up; however, these rates are much lower in patients infected with non-A HBV genotypes. Continuous treatment with second generation NUCs, entecavir or tenofovir disoproxil fumarate (TDF) for up to 10 years, results in overall HBsAg seroclearance rates of 0-5%, with higher rates in HBeAg-positive patients. HBV genotype is the strongest predictor of HBsAg seroclearance in patients who received PEG-IFNα treatment while low baseline HBsAg level is the best predictor of HBsAg seroclearance in patients receiving NUC treatment.[26]
Novel HBV therapeutics (Table 3)
Current antiviral therapy for chronic hepatitis B is associated with low rates of HBsAg clearance; thus, long-term treatment is required to maintain virologic response. Therefore, there is great interest in developing finite treatment strategies that will achieve durable off-treatment clearance of circulating HBsAg and HBV DNA. This will require therapeutic strategies that target the high viral and antigen burden as well as inadequate host immune responses.
Antiviral strategies
Currently the only approved direct-acting antivirals are NUCs. The HBV lifecycle provides many other targets for new molecular entities, including HBV entry, core protein assembly, viral protein synthesis and virion/subviral particle assembly and release (Fig. 1).
Entry inhibitors: bulevirtide, a small myristoylated synthetic lipopeptide corresponding to the HBV preS1 sequence, blocks the binding of HBsAg to NTCP (sodium taurocholate co-transporting polypeptide), the entry receptor for both HBV and HDV. Bulevirtide’s antiviral effect is mediated via prevention of infection of uninfected hepatocytes, but it may also block entry of new virions to infected hepatocytes. Most studies of bulevirtide have focused on chronic HDV infection where 24 weeks of bulevirtide monotherapy has been shown to result in ≥2 log reductions in HDV RNA levels in 46-77% patients at the end of treatment, though viral relapse was observed in most patients during post-treatment follow-up. Decline in HBsAg level by ≥1 log was rarely observed during bulevirtide monotherapy, but was more common with a combination of bulevirtide and PEG-IFNα though HBsAg clearance remained a rare event.[41]
Inhibitors of core synthesis/capsid assembly modulators: the HBV core or capsid protein plays multiple essential roles in the HBV lifecycle and is an attractive target for small molecule inhibitors. The primary mechanism of action of core inhibitors or capsid assembly modulators (CAMs) is inhibition of HBV replication by interfering with HBV capsid assembly and encapsidation of pregenomic RNA (pgRNA). Secondary mechanisms of action of CAMs are inhibition of cccDNA establishment by interfering with capsid disassembly and inhibition of replenishment of cccDNA by interfering with intracellular recycling of HBV nucleocapsids.
The first CAM to enter clinical trials, NVR3-778, had a modest antiviral effect. [42] Other CAMs (JNJ-379, RO7049389, vebicorvir and ABI-2158) are more potent and reduce HBV DNA by 3 logs and HBV RNA by 2 logs after 28 days. Although modest (<0.5 log) reductions in HBsAg, HBeAg and hepatitis B core-related antigen (HBcrAg) levels were observed after 24 weeks of NUC+CAM in treatment-naïve HBeAg-positive patients, these were not different from those who received NUC monotherapy.
The combination of a CAM and a NUC may have synergistic antiviral activities. A pilot study of patients on vebicorvir plus entecavir showed that combination therapy resulted in more rapid and more marked decline in HBV DNA and HBV RNA levels compared to entecavir monotherapy, but despite having undetectable HBV DNA and HBV RNA levels, all patients rapidly relapsed when treatment was stopped at week 48.
For CAMs to become an integral part of finite HBV functional cure strategies, they will need to contribute to HBsAg loss through their secondary mechanism of action. The next-generation CAMs entering clinical development (ABI-H3733, AB-836, ALG-000184, VNRX-9945) are more potent and pharmacokinetic studies suggest that the hepatocyte concentration necessary for their secondary actions can be achieved.
There is emerging evidence that CAMs may be associated with liver-related toxicity. Several CAMs have been associated with treatment-emergent Grade 2-4 ALT elevations.
Although RO7049389-induced ALT elevations are thought to be immune-mediated “good flares”, ALT elevations induced by other CAMs reflect drug-induced liver injury and their development has been halted. Naturally occurring HBV variants resistant to CAMs have been identified and monotherapy with CAMs has been reported to lead to the selection of resistant HBV variants and virologic breakthrough.
Thus, CAMs must be used as part of a combination therapy regimen and pre-existing CAM-resistant HBV variants may reduce the efficacy of some CAMs.
Inhibitors of HBV protein synthesis/translation inhibitors: Translation inhibitors that silence HBV RNA could contribute to HBV cure by inhibiting virion and subviral particle production, thereby boosting host innate and HBV-specific immune responses. Strategies to inhibit translation include small-interfering RNAs (siRNAs) and antisense oligonucleotides (ASOs), which have very different pharmacodynamic properties, reflecting differences in their chemistry, cellular delivery and intracellular handling. siRNAs are duplex (double-stranded) RNAs with guide (23 nucleotide) and passenger (21 nucleotide) RNA strands, whilst ASOs are single-stranded DNAs (8-10 nucleotides). “Naked” ASOs can readily enter all cells, but siRNAs must be GalNAc conjugated (conjugation with N-acetylgalactosamine to bind to asialoglycoprotein receptors on hepatocytes) to ensure delivery to the hepatocyte. siRNAs accumulate within endosomes and form a stable guide RNA-argonaut (AGO) complex, that can cleave multiple target HBV mRNAs, thereby amplifying gene silencing and permitting infrequent dosing (monthly or less). In contrast, ASO binds to target HBV mRNA sequences, forming a DNA-RNA hybrid that is rapidly degraded by cytoplasmic RNase-H, necessitating frequent dosing (weekly or more).
The first siRNA, AR520, was designed to trigger at the common 3’ termini of HBV RNAs (pregenomic and messenger RNAs). While it led to marked decreases in HBsAg levels in HBeAg-positive patients, the effect was diminished in HBeAg-negative patients. Subsequent studies revealed that the trigger site is often missing in HBeAg-negative patients in whom the source of HBsAg is predominantly from integrated HBV DNA.
New siRNAs and ASOs target overlapping regions of the HBV genome or have multiple triggers to improve efficacy.
The GalNAc-conjugated siRNAs in development (JNJ-3989; VIR-2218; RG-6346; AB-729) are safe and associated with predictable on-treatment HBsAg declines, durable for many months post-treatment.
HBsAg kinetics from phase I studies predicted that monthly siRNA dosing would lead to HBsAg loss after 12 months. However, phase II studies demonstrated plateauing of HBsAg after 16-20 weeks without HBsAg loss, despite continued administration for up to 48 weeks.
Nonetheless, even in patients who received only 2-3 doses, decreased HBsAg levels were maintained for up to 9 months. The absence of ALT flares in the siRNA studies cast doubt on whether profound HBsAg reduction would restore host immune responses. However, a recent study of multiple doses of AB-729 observed that rapid HBsAg reduction was associated with moderate ALT elevations and upregulation of HBV-specific T-cell activation markers in some patients.
This suggests that efficacy of siRNA therapy may be enhanced by combining with immunomodulatory agents, a hypothesis that is supported by profound HBsAg reductions associated with ALT elevations in a study combining an siRNA with PEG-IFNα.
The only naked ASO in clinical development (GSK836) led to rapid and profound HBsAg reductions (3-4 log reduction within 28 days) followed by ALT elevations, thought to represent HBV-specific immune reconstitution.
Four patients in that study lost HBsAg by Day 29 and 2 remained HBsAg negative for more than 3 months. Additional studies of GSK836 and GalNAc-conjugated ASOs with enhanced antiviral efficacy are ongoing.
cccDNA targeting: Complete or sterilising cure will require approaches that degrade or silence cccDNA. Direct cccDNA silencing: In in vitro experiments, combining Cas9 with guide RNAs which target conserved HBV sequences leads to profound reductions in cccDNA and all HBV proteins.
The main challenges of translating gene editing to the clinic will be the need to achieve 100% efficiency of delivery to infected hepatocytes and to mitigate the long-term risks of possible chromosomal translocation following editing of integrated HBV sequences. Indirect cccDNA epigenetic silencing: small molecules that target host histone deacetylases, acetyltransferases and demethylases can also interrupt the normal epigenetic regulation of HBV gene expression thereby silencing cccDNA transcription. The oral (H3K4me3:H3) KDM5 demethylase inhibitor demonstrated potent reductions of HBV antigen production in vitro but clinical development was stopped because of safety concerns.
EYP001, a farnesoid X receptor agonist, is a potent inhibitor of cccDNA transcription in vitro and is in clinical development.
The HBx protein modifies the epigenetic regulation of cccDNA function. Small molecules, siRNAs and gene editing approaches to knockdown HBx are all being evaluated in preclinical studies.
Inhibitors of HBsAg secretion (NAPs and STOPs): Nucleic acid polymers (NAPs) have been reported to rapidly reduce circulating HBsAg with minimal effect on other viral markers, when given as either monotherapy or in combination with IFNα.
In a phase II trial of 40 patients, TDF for 24 weeks followed by the addition of PEG-IFNα and REP 2139 or REP 2165 for 48 weeks, resulted in 44% HBsAg loss at the end of treatment, and 39% of patients met criteria for functional cure >24 weeks after the end of treatment. The frequent ALT flares suggested hepatotoxicity from intracytoplasmic accumulation of HBsAg. However, in vitro studies have demonstrated that NAPs enhance intracellular degradation of HBsAg via proteasomal and lysosomal degradation. These encouraging results need to be confirmed by larger multicentre studies.
S-antigen traffic-inhibiting oligonucleotide polymers (STOPs) are a new class of oligonucleotides that share structural similarity with NAPs but have novel chemical features which enhance antiviral activity and allow for subcutaneous administration, in contrast to NAPs which require intravenous administration. ALG-010133 is the first STOP in clinical development.
Host-targeting agents: Inarigivir activates the cellular retinoic acid-inducible gene I pathway, thereby inhibiting HBV polymerase and inducing endogenous IFN induction. Although early studies reported a dose-related reduction in HBV DNA levels, higher doses resulted in severe hepatotoxicity, with multiple cases of acute liver failure and one death, leading to its discontinuation.
Other classes of antivirals that target host pathways include transcription inhibitors (KDM5 demethylase inhibitors, FXR agonists) and mRNA destabilisers (dihydroquinolizinone compounds, PPAD-5/7 inhibitors). Development of these classes of drugs has been hampered by off-target toxicities.
Immunomodulatory therapies
Patients with chronic HBV infection have impaired immune responses to HBV.
HBV-specific T cells are decreased in number and functionally defective with features of exhaustion as exemplified by upregulation of multiple co-inhibitory receptors, and transcriptional, metabolic, and epigenetic defects. Restoration of immune response is important in achieving functional cure to eliminate infected hepatocytes and/or block infection of new hepatocytes. Studies of patients who achieved HBeAg or HBsAg seroclearance after NUC or PEG-IFNα treatment showed that HBV-specific immune responses can be restored, suggesting that HBV-specific T cells are exhausted but not deleted.
Multiple approaches to stimulate or to remove blockade of HBV-specific immune responses have been studied, but results have been poor (Fig. 2). Several factors might have contributed to failure of these early attempts: persistent high levels of circulating HBsAg, use of HBV surface antigens/epitopes that patients are tolerant to, and reliance on single (in contrast to multi-prong) or subdued (for fear of severe flare if rigorous immune response is achieved) approaches.
Checkpoint inhibitors: In vitro studies showed that programmed death receptor 1 (PD-1) blockade alone is not sufficient to completely reverse the immune function impairment that is characteristic of chronic HBV infection. Studies in woodchucks showed that the combination of PD-L1 blockade with a NUC and therapeutic DNA vaccination may be more effective.
A phase Ib study of 12 weeks treatment with nivolumab, a PD-1 inhibitor, with or without GS-4774, a yeast-based therapeutic T-cell vaccine that expresses HBV core, envelope and X antigens found that 3 of 22 patients in the high-dose group had ≥0.5 log decrease in HBsAg at week 24 with 1 patient having undetectable HBsAg that persisted at 12 months post-treatment.
Improvement in HBV-specific T cell responses was observed in some but not all patients, with the most marked improvement in the patient who achieved HBsAg seroclearance.
Metabolic modulation: T-cell activation and function require dynamic metabolic adaptations. T cells in patients with chronic HBV infection have been shown to be deficient in arginine, and partial restoration of CD8+ T cell function may be restored by in vitro arginine replenishment.
HBV-specific T cells in patients with chronic HBV infection also exhibit mitochondrial dysfunction. A recent study found that mitochondria-targeted antioxidant therapy can improve HBV-specific T cell function in vitro.
Whether these approaches will restore HBV-specific T cell function in patients with chronic HBV infection remains to be determined.
Therapeutic vaccines: Vaccines to boost HBV-specific T or B cell immune response might be needed to achieve functional cure. Multiple approaches, including with protein-, peptide-, DNA- and viral vector-based vaccines have been evaluated with very little success to date. Two recent studies using GS-4774, a yeast-based vaccine, in NUC-naïve and NUC-experienced virally suppressed patients, showed minimal reduction in serum HBsAg levels despite increased production of cytokines, such as IFNγ, by HBV-specific CD8 T cells.
Recent approaches to therapeutic vaccines have focused on using viral vectors, inclusion of antigens from other regions of HBV (such as core and polymerase instead of surface) or modifications of the HBs epitope that appear to be better recognised by patients who spontaneously cleared HBsAg, as well as the addition of check-point inhibitors or metabolic modulators to overcome HBV-specific immune exhaustion. Interest in the viral vector approach has been spurred by COVID-19, for which several vaccines were developed using adenoviral vectors. The ChAdOx-HBV vaccine in combination with a PD-1 blocking monoclonal antibody will be tested in a phase I/IIa trial.
Data on many of these new approaches are preliminary and some have only been tested in animal models or healthy volunteers; thus, the efficacy of these new vaccines in patients with chronic HBV infection remains to be seen.
Chronic exposure to high circulating levels of HBsAg has been identified to be a major contributor to HBV-specific immune exhaustion. Thus, most experts have agreed that in addition to suppression of HBV DNA replication, a decrease in HBsAg production and/or release would be important to increase the efficacy of therapeutic vaccines.
Adoptive transfer of genetically engineered T cells: Transplantation of bone marrow from persons who have spontaneously recovered from HBV infection have been shown to result in HBsAg clearance in patients with chronic HBV infection demonstrating that transfer of HBV-specific T cells can achieve functional HBV cure.
Several strategies to transfer genetically engineered T cells to restore HBV-specific immune response have been explored. Re-direction of patient T cells by transfer of HBV T-cell receptor genes has been tried in humans with HBV-related HCC and appeared to be safe and effective.
CAR (chimeric antigen receptor) T-cell therapies, which are increasingly being used for haematological malignancies, have been proposed, but genetically engineering T cells is technically complex and unlikely to be widely adopted as a treatment for chronic HBV infection.
Stimulation of innate immune response: Innate immune response mediators may enhance antiviral immunity, but they lack precision. An oral toll-like receptor 7 (TLR7) agonist, GS9620, resulted in a sustained decrease in serum HBV DNA and HBsAg in chimpanzees with chronic HBV infection.
However, a phase IIb trial in humans virally suppressed on NUC showed minimal decrease in HBsAg levels despite increased cytokine production by T cells and natural killer cell activation.
Similarly, a recent trial of the TLR8 agonist selgantolimod, in combination with a NUC, resulted in minimal reduction in HBsAg levels.
Inarigivir, which induces the intracellular IFN signalling pathways showed promise in early clinical studies, but development was terminated following the death of a trial participant.
Stimulation of B-cell immune response: The importance of B-cell immune responses is evident from the high rate of HBV reactivation in patients receiving B cell depleting (anti-CD20) therapies. HBsAg-specific B cells persist in many patients with chronic HBV infection, albeit at low frequencies and functionally impaired.
Boosting endogenous B cell responses has several potential advantages beyond increasing antibody production; however, an effective strategy remains elusive and will succeed only if HBsAg production is markedly suppressed. Infusion of antibodies to HBsAg may neutralise circulating HBsAg and prevent new infection of hepatocytes but the effects would be short-lived.
Combination therapy – rational approach to combining treatments
Why is combination needed?
Multiple steps need to be satisfied to achieve the goal of functional HBV cure: i) complete suppression of HBV DNA replication, ii) inhibition of HBsAg production from both cccDNA and integrated HBV DNA, and iii) restoration of host innate and HBV-specific immune responses (Fig. 3). While one or two classes of drugs might suffice in achieving functional HBV cure in a small percentage of patients, the combination of several classes of drugs will be necessary to achieve this goal in a high percentage of patients.
Recent studies showed that blood from patients with undetectable serum HBV DNA on NUC therapy contain residual virus and could transmit HBV.
[74]
One study showed that the combination of a CAM and a NUC can lead to more rapid and marked suppression of HBV DNA as well as HBV RNA, but the effect on HBsAg level was small. The addition of entry inhibitors may contribute to suppression of HBV DNA replication. siRNAs and ASOs can result in a rapid decrease in HBsAg production but HBsAg clearance is rare. The addition of PEG-IFNα after HBsAg and HBV DNA have been suppressed to low levels might increase the likelihood of HBsAg seroclearance. Sustained functional cure will require restoration of HBV-specific immune responses. For some patients, this might be accomplished after complete suppression of HBV DNA replication and inhibition of HBsAg production but, for many, immunomodulatory therapies will be necessary.